Possibly one of the most pressing issues right now is how automation might (keyword: might) tear apart the economy. Many studies have arisen showing that automation is going to be take out millions of manufacturing jobs globally. In response, even people on the right such as paleoconservative pundit Tucker Carlson and libertarian political scientist Charles Murray have come out saying we need to either heavily regulate it or find ways for the employed to be secure. The issue has started to bring people together.
I remember reading Economics in One Lesson by Henry Hazlitt and even when he gave a rebuttal to the claim automation will cause a massive wave of unemployment, I was hesitant to believe him. His explanation relied on long term jobs staying in place that reality tells us wouldn’t be. He also missed the fact that manufacturing jobs are largely held by people that haven’t gone to college or don’t have a formal education. The jobs he wants to replace them would require a higher level of skill and education.
Hazlitt isn’t alone in this view. An article from Concept Systems tells us automation is actually great for the economy! The main arguments for this are along the same ideas. The business owners now have more money to invest into workers and becoming more efficient. They also become more competitive and need to increase productivity and therefore will eventually hire more workers (supposedly). He also makes note of the same argument that jobs will shift because people need to build and maintain these machines, stating,
“In terms of the people creating and maintaining the machines, their skills will be highly valued. More people will study engineering and robotics, and skilled jobs will steadily grow in demand. Even what could be considered unskilled jobs, or assembly of machines or routine maintenance, will need to be completed by humans.”
Hazlitt and other Austrian economists warn us to stop looking at individual fields and that the short term doesn’t matter. The argument made is this tells us nothing about the economy in the long run. Where one job leaves, more money is saved and invested into another job or invested in the same job down the line. This line of thinking is completely okay to libertarians. But, it’s also a little sociopathic.
A great study came out by self-described libertarian, Johnathan Haidt in 2012 that would explain this. In the study, he looked into the personality traits of others who called themselves libertarian in order to associate a psychological archetype to the ideology. His results shouldn’t surprise us. Libertarians are often amoral or lack moral conviction at the very least. They also have less aversion to hedonism and less social interdependence and relatedness than even liberals. The libertarian personality is not one of liberal sympathy nor one of conservative moral conviction.
While libertarians are supposedly individualists who care about the”individual above all,” they view the market as a whole collective – a singular working entity and if some areas are lost from it, they may be replaced – emotions aren’t involved whatsoever. With automation, the large transferal of jobs is much less important. As long as the market stays intact right? People like Tucker Carlson, on the other hand, are much more concerned about the psychological toll this will have on the working class. He is concerned about the average Joe who is going to take a lot of damage from losing his manufacturing job.
This isn’t to claim Carlson’s position is perfect. In his conversation with Ben Shapiro, he sounds almost Machiavellian in his methods of implementing a ban on automation. Tucker says he “would make up a reason” in order to get it passed. It’s not like politicians don’t lie already but, I guess hearing him say that sent me back a bit. Furthermore, the automation ban could prevent a lot of great improvements being made in the future. But the question here is the effect it will have, even in the short run. The long term effects, as made to be so important by Hazlitt, might be minuscule, hard to measure, non existent, or only as a consequence from a much harmful means. In addition, as Alan Keynes said, “In the long term, we are all dead.”
Regardless of this, I still think Carlson is quite a bit wrong.
First, the effects of automation are completely blown out of proportion. A recent study from 2018 analyzes the actual causes of the recent losses in manufacturing jobs. The current major belief is that automation is the biggest cause for loss in this area, but as the paper points out automation actually has very little causal relation with manufacturing employment rates over recent years. The real causes, as the author finds, are major increases in productivity as well as liberal trade policies, particularly with China.
This doesn’t go towards Hazlitt’s argument, either. For one, Hazlitt makes this major case that productivity is what matters in a market place, not unemployment. But, this major increase in productivity has led to unemployment without any benefit for those now unemployed. In a situation where Hazlitt’s makes sense, sure, there would be high unemployment, but only because productivity is high enough people can stay home. This is what allows for the stay-at-home mom. But in this case, productivity is increasing and it is just leading to people being unemployed.
If productivity was increased and people became unemployed you would need one of two things to happen. One method could be an income (possibly a UBI; we’ll discuss this later) that supports those unemployed. Now that productivity is up, less jobs are needed but the people who become unemployed won’t gain anything. The other way this could be helped is if productivity leads to an increase in wages. With that increase in wages, the workers still employed could go and invest in the market and create new jobs. But, productivity has been increasing for years and wages have stagnated. If we want the same people who lost their jobs to be re-employed, they’ll need other skills.
Secondly, the paper argues trade with China seems to have an effect as well. This is something Hazlitt was in favor of, at least free trade that is. Hazlitt and other Austrian economists are not being helped by the fact that liberalization of trade has caused a loss of manufacturing jobs.
The estimated manufacturing decline for the next ten years in America is predicted, at least by DataUSA, to be a 0.7% drop. This comes out to about a loss of 89,250 jobs. Seems like a pretty big job loss, although maybe take the Austrian theory (that the jobs will just shift elsewhere) and we’ll gain some of those jobs back. But, once again, viewing the market size as a collective is not always the best strategy.
The Verge released an article in 2017, titled “Automation threatens 800 million jobs [worldwide], but could still save us, says report”. They cite a fairly popular study that came out from the McKinsey Global Institute which shows how automation will destroy jobs worldwide. But, they note, the report also states it doesn’t take into account the jobs being created. The author of the article writes,
“The changes won’t hit everyone equally. Only 5 percent of current occupations stand to be completely automated if today’s cutting-edge technology is widely adopted, while in 60 percent of jobs, one-third of activities will be automated. Quoting a US government commission from the 1960s on the same topic, McKinsey’s researchers summarize: “technology destroys jobs, but not work.” As an example, it examines the effect of the personal computer in the US since 1980, finding that the invention led to the creation of 18.5 million new jobs, even when accounting for jobs lost.”
Essentially, the idea of automation coming through and leaving everyone hopeless isn’t necessarily true and, in fact, we may have a large increase in jobs because of it.
The effects of automation are made more pleasing for us by Timothy B. Lee in an article for Vox. His major argument is the increase of automation is actually going to make human labor a lot more valuable. In the beginning of an article (which is actually an excerpt from a book), he argues while manufacturing jobs are being taken out by automation, “the demand for labor-intensive services is soaring. He points to the rise of Etsy for example, which has grown immensely over the last few years and is completely based off personally made products or items that aren’t mass produced.
Even in restaurants, which are one of the fields most affected by automation, there is a great demand for human labor. In Starbucks, people become dissatisfied with using a machine to make their order and overall, not having that human interactions. Lee writes people don’t just go to Starbucks for a cup of coffee, but “[p]eople go to Starbucks because they enjoy the experience. And that experience has an important performative dimension — customers want to feel like their barista devoted personal attention to preparing their cup.” Humans can be trained to make connections with other humans; robots, at least until AI takes us downhill, can not.
Wineries and breweries are not much different, according to Lee. With the rise of automation, we’ve still experienced a massive increase and wineries and breweries across the nation. These are, once again, an experience created through human effort, not blatant machinery. As more fields become more automated, the general desire, and in turn necessity for human labor becomes greater. The value of said labor becomes more important and even increases in monetary worth. From this position, automation may not only take away our fears of job loss, but of the growing issue of consumerism.
In regards to strictly automation, we have seen this in real life through the ATM effect. When the automated teller machines (or ATMs) started coming about, a lot of people were afraid they were going to replace bank tellers. But, ATMs couldn’t complete every single task and instead of replacing teller’s jobs, it just changed them. While ATMs took over specific, easy roles, certain jobs were still needed and they actually made bank tellers more valuable because of it.
JP Morgan has conducted a study on how automation will shift the economy in a good direction. I read an article about it from Business Insider and it was incredibly interesting, especially when it contrasts the views of so many on the left or populist right. The article said about the study,
“”In the past, technological innovation transformed society and increased labor productivity in three key ways,” the report states.
- “Replacing existing workers with machines, and thus producing at least the same output with fewer workers (e.g., refrigeration vs. the ice man);”
- “Complementing existing workers’ jobs, boosting output per worker by automating some of their tasks (e.g., power tools);”
- “Creating entirely new, higher productivity industries (e.g., computer software engineering), offsetting the displacement of workers by machines, or replacing altogether industries that have been made obsolete.”
The extra growth delivered, the report said, would likely be in the order of around a 1%-1.5% boost to global GDP. The most recent estimates suggest that this is around $75.6 trillion, meaning that any boost would be worth in excess of $1 trillion. That’s roughly equivalent to the GDP of Mexico or Australia in 2016.”
So with all this on the side of automation, it’s almost impossible for me to hop on Tucker Carlson’s train, at least all the way. Conversely, it’s not easy to support a laissez-faire approach to this either. Regardless of the benefits we see here, there will be a lot of pain in the process, that is, if we don’t find solutions to amend said pain. The best solution, at least the way I see it, is to allow a good amount of automation but first, organize solutions for the decreases in jobs that is to come. Put simply, there needs to be a good way to adapt to upcoming circumstances.
One of the most interesting studies I read came from the left-wing Brookings Institute. It is titled “Automation could create a more inclusive economy through high-skilled side work” and was written by Gregory Ferenstein. Basically what Ferenstein outlines is a bit of research he’s done that shows people with a side job largely face pay increases after automation takes place. He places emphasis on the necessity for people in jobs that may be replaced to pick up gig work as extra job training.
The article starts off with a story about a guy the author met who works for Ford. The employee took online classes to pick up new skills over time. What this allows him is access to new careers if his is replaced by automation. He says, “If Ford calls Today and says ‘we’re closed’, I’ll be good tomorrow.”
Then, we move into actual data on the matter. Ferenstein utilizes a number of data sets including the Census Bureau to measure his theory. He finds the 75th percentile of employees with two jobs and who had theirs displaced found themselves making 19% more per week than before and the bottom 25th percent facing a bit worse of a scenario, retaining only about 59% of their previous income. People with graduate degrees do significantly better, earning 62% more than those without. He claims this as possible evidence people with higher skill ranges are “probably leveraging second jobs for upward mobility.”
The datasets he uses afterwards are even more promising; Ferenstein states,
“One of the richest datasets comes from Britain’s Understanding Society, which has exhaustively tracked the economic cohorts of Brits over several years, and does perhaps the best job of asking about multiple income streams. From this dataset, I looked at how people with a skilled side job differed from single jobholders who switched occupations. People with a skilled side job end up with a median income retention of 111 percent of previous earnings (with the 25th percentile retaining 67 percent and the top 75th making 295 percent).
The Understanding Society sample is much smaller than a less-detailed monthly supplement to The Current Population Survey, the Outgoing Rotation survey. I compared my findings with American data and found that people holding a skilled side job, defined as one that pays above-average wages, retained 97.7 percent of income after changing occupations, with the 25th percentile retaining 60 percent and the top 75th earning 168 percent.”
Another company called Talent Works skimmed through resumes and job offers so he could see the effect having more skills had on peoples wages. Talent Works found 5 years of part time experience would increase wages 54%, fighting back against the dip in wages caused by automation.
The biggest thing, Ferenstein says, is just making it acceptable to work more than one job in the work week. He says worker displacement is not actually that big of an issue if we can train our populace well enough to be secure once the automation hits them. This will be great for a lot of people when mixed with the concept automation increases value of human labor. Many people might go into self employment and do just fine. Automation can bring some great things here; we just need to adjust for it.
How do people get these skilled side jobs and gigs? Well, through education of course. This brings us to another method of trying to secure people in the case of worker displacement through automation. We need to allow people easy education that prepares them not only for the workforce, but for the case they get screwed. But, saying “they need better education” is not a pleasing answer. This gives a “solution” with no meat or “how-to”. Saying “healthcare needs to be fixed” may be agreed upon until we talk about how.
An online acquaintance of mine, Chris DeWitt, wrote a great article for The Centre Right Review that covers one of the major issues developing wealth inequality – a race between education and technology. What happens, according to a paper by Lawrence Katz and Claudia Goldin, is jobs pay more to those with more advanced skills and the change in technology over the years creates an increase in skill biased careers. This is clearly applicable to the automation issue, regarding low skill manufacturing jobs and high skill computer/tech jobs.
Chris DeWitt overviews methods to try and reform education in an effort to catch up to technology. After pointing out the flaws with a socialized college system, he explains,
“So policies should be taken to lower the costs of colleges and subsidize fields that are needed most. Grants, federal aid, and other forms of government assistance would help make college more accessible to more people and help close the gap between education and technology.”
Finally, he cites a Brookings Institute study regarding success in life, offering the option of encouraging personal responsibility so people can get to college and pick up advanced skills to prepare them for life. DeWitt states,
“Finally, to wrap up the education section we should look at a study conducted by the Brookings Institute. The Brookings Institute found that just following 3 simple rules would help keep children from poor families out of poverty. If children at least graduate high school, have a full time job, and wait until they are 21 to get married and have children there will be a 98% chance they will not be in poverty, and nearly 75% have joined the middle class defined as earning at least $55,000 a year (Brookings, 2013). Although other factors are certainly at play, simply instilling these values in children by policy makers and schools would help to close the gap and make everyone better off.”
America has gone through phases similar to this before. An Instagram user (@americles) explained this to me through a historical lens. Essentially, automation will be to manufacturing as manufacturing was to agriculture. Agriculture used to be roughly 60% of the economy. People were afraid the economy was going to collapse with the rise of manufacturing, because it required more knowledge and more skills than agriculture did. At the time, though, the high school movement began and it pushed for education and prepared people to work in more skilled jobs. This allowed a smooth transition. Agriculture went from 60% to 2% of the jobs, but the economy maintained its well being.
This pushes a couple important points. For one, we are able to get through these issues seamlessly. The hype about how terrible it is only applies if we don’t prepare for the worst. What happened with agriculture and manufacturing was made possible through better educational opportunities and schools working more to prepare people for the workforce. The issue does seem to be, as Chris DeWitt says, a race between education and technology. Once we can start training our populace for these high skill jobs, we can beat automation.
Regarding preparing ourselves for automation, I think we should address the elephant in the room – Universal Basic Income (or UBI). Many on the right are hopping on the Yang Train because he is the only Democratic politician openly advocating for a universal basic income (as well as the only one not making their campaign about being anti-Trump, although I’d say Bernie is not bad in that either) and because a lot of people, besides right-libertarians or those who have actually read about the effects, think it is the only solution to the rise of automation. Despite his incredibly socially liberal policies, wanting to make Puerto Rico a state, preventing Republicans from ever getting a vote, and the fact that he will nominate liberal judges, somehow right wingers are still hopping on the Yang Train.
One goal from some of the far right is accelerationism. The idea behind accelerationism was originally to expand capitalism and cause a collapse so Marxism could become viable. Of course, some on the reactionary right have made it more about increasing immigration and socially liberal policies while increasing the expenses America is paying per year to cause the collapse. Of course, I have some major issues with this idea though.
For one, it’s likely not going to go in their favor.
When the Italian communist Antonio Gramsci moved to Soviet Russia, he took notes of Lenin’s regime and why it failed. Gramsci wrote the only way to destabilize capitalism in the West was to destroy the systems that held it up – one being Christianity. Along with Christianity comes traditional gender roles, capitalism, and a respect for order and hierarchy that did not allow the Marxist system to come into the world naturally. He found the method devised by Marx in the phrase, “Workers of the world unite!” can only come second to destroying the culture. Gramsci writes about Russia,
“… the state was everything, civil society was primordial… in the West there was a proper relation between the state and civil society, and when the state trembled a sturdy structure of civil society was at once revealed. The State [in the West] was only the outer ditch, behind which there stood a powerful system of fortresses and earthworks.”
When referring to accelerationism from the right, we are typically referring to liberal policies. This is clearly an issue – once we implement socially liberal policies and allow the situation where America is allowed to collapse, the idea of a Marxist regime is completely feasible. People in America have already started losing religiosity and have been deconstructing the “oppressive” hierarchies we had built up traditionally. Sure, groups like the Alt Right and white nationalism have been growing over the last decade. So, have groups like communists – maybe more so. When the Overton Window in America has moved further left and socially liberal policies have been implemented, who wins that war between communists and the far right?
Furthermore, if they are on the reactionary right, they want an American empire, right? Well, you aren’t going to get that once you’ve collapsed America as it is. America is currently the world’s biggest empire but China is catching up quickly. Do you really think they’re going to allow us to expand and build up power when they’re in charge? By the time we’ve balkanized or collapsed, China is going to rise into power, they’re going to seize assets because we already owe them a ton in debt, and they will probably use military force against us.
(For more regarding this topic, you can refer to my article, The New American Revolution.)
This idea of accelerationism is going to cause a lot of despair, especially to the elderly. Your best solution, while it sounds “lame and blue-pilled” is to try and fix what we have, raise birth rates, and use better methods to tackle automation.
This is not to completely disavow the idea of a UBI. In fact, I’ve kind of avoided fully giving an argument on it until this point. A universal basic income is not something I’m against. At the moment, it’s not something I’m completely for, either. It doesn’t seem to increase employment. It’s also not as expensive as made out to be given you update the tax structure. While the economic effects aren’t huge, there are some good psychological effects, too.
Earlier this year, I read a good article from the Huffington Post regarding how the test of Universal Basic Income worked in Finland. What they did to test it was find 2,000 unemployed Finns and start giving them $640 a month for two years. Those involved in the experiment had no complaints with that. The trial did not increase employment significantly (at least for the first year). But, as the article says there were great psychological improvements for the participants. Their personal well-being grew as they had the assurance of being safe in case anything happened, regarding hours worked or losing their jobs.
“Constant stress and financial stress for the long term – it’s unbearable. And when we give money to people once a month they know what they are going to get,” said Ylikännö. “It was just €560 a month, but it gives you certainty, and certainty about the future is always a fundamental thing about well being.”
A couple of articles place the blame for “failure” of these programs (or at least failure towards the supposed goal of them) in the areas they’ve been tested, not on the programs themselves, but on issues regarding external circumstances. An opinion piece for The New York Times is titled, “Universal Basic Income Didn’t Fail in Finland. Finland Failed It”. Another article for Technology Review outlines three major, consistent problems with implementations of Universal Basic Income. These were the external politics of the country, the funding given for these projects, and how they disrupt currently existing social benefits.
In terms of the cost, many tout a $3-4 trillion price tag. Though this is based off a misconception on how it would be implemented. Karl Widerquist points out this is a gross cost estimate instead of a net cost estimate which isn’t exactly what we want for a universal basic income because not only is it redistributive, but it gives back to those that pay into it (aka, the wealthy). So Widerquist says,
“Suppose the government gives $1 to Haveless, financing it by taking $1 in taxes from Havemore. The benefit of this targeted program is that Haveless gets $1. The cost is that Havemore has $1 less. This program effectively makes Havemore give Haveless $1. Next, the government universalizes this program by making Havemore give $1 to himself. How much more does this cost Havemore? Nothing. The cost and benefit cancel each other out, because they go to the same person at the same time in the same form. There may be a small administrative cost of running the program (see below) but the $1 that Havemore gives to himself indicates nothing whatsoever about the real redistributive burden or feasibility constraints of this transfer.
It costs you something if the government takes $1 from you and gives it to someone else, but it doesn’t cost anything if the government gives and takes the same $1 from the same person at the same time. There is no economic or financial limit to how much money the government can create out of thin air and give to you and everyone else, as long as it immediately taxes it back. In financial terms, this giving and taking-back makes up the bulk of what UBI does, but it is not a cost to taxpayers, to the government, or to the economy, and cannot truthfully be counted as a cost at all.”
The actual cost of a UBI is not 3 or 4 trillion but around $500-600 billion (at least at a top tax rate of 50%), about a third of what we pay for all entitlements right now.
In addition, it appears universal basic income would grow the economy largely. A study by the Roosevelt Institute found a universal basic income would grow the economy by $2.5 trillion, the increase becoming larger with the size of the UBI (although I would bet there’s a Laffer curve there). The only dispute I’ve seen of this study comes from the libertarian thinktank, Foundation for Economic Education (FEE). But, their dispute is based on a price claim and a couple hypotheticals that come with Austrian thinking. The refutation lacks much substance and tries to say “Well the Roosevelt study was based off of hypotheticals” when that’s most of economics anyways, as well as most of their ideology. It’s a pretty sad excuse for a rebuttal.
It does make sense the economy would grow with a universal basic income. After all, poor people spend a much larger percentage of their money than rich people do. Having more disposable income means more goods are bought and in turn, more jobs are created. We’ve seen in Finland at least, employment doesn’t go down when you institute a UBI, meaning this whole idea of “People will be lazy!” is just as untrue as “More people will be employed!” It would also allow for a repeal of the minimum wage (if you wanted). This would give more profits back to businesses, especially smaller business, which would potentially increase their spending into resources and innovation and make them more competitive. It’s not like people would go without pay either, at least in theory, because businesses now have incentive to keep their workers in and possibly increase the wages they were paying them anyways.
I think the psychological effect is going to be one of the biggest hitting points. If all people working in manufacturing had this extra sense of hope that at least some of their income is secured when a robot comes in to replace them, I’m sure a lot of them would report higher well-being. Providing an extra $1000 a month anyways is great for allowing people to work on personal projects, start saving for their children, or keep up on their bills. It would be useful for an already crumbling working class and potentially help with the major issues affecting them right now.
Overall, a universal basic income, if it can be done correctly, could have some major benefits. While I don’t think it’s necessarily the best option to address automation, or at least it’s not the first and most vital option, it might be useful to have in the future. It would also cut down on government bureaucracy in the welfare system and make it a lot simpler. This makes government more efficient and smaller, hence why some libertarian thinkers such as Milton Friedman have supported it in the past.
For people working in manufacturing careers right now – your job isn’t as threatened by automation as you might think. This is true for a few reasons. Automation is hardly to blame for the job losses in the last few decades and in some cases where automation takes reign, jobs are kept (such as with ATMs). But that being said, your job is still at risk. Liberal trade policies such as the WTO have allowed them to be shipped offshore and massive increases in productivity without any substantial increase in wages has become a major issue. It’s not like automation is going to stagnate or grow at the same rate. It may grow exponentially. Twenty years from now, it may have more capability to replace your job than it does now.
You do need to secure your job and we as a society need to help this happen. Plenty of online college classes are available as well as great training courses through Lynda, Skillshare, and Coursera, all websites I’ve used in the past and enjoyed. Picking up part-time work and getting an extra few hours a week will not only give you a little bit of dough, but it will help you build up new skills. If you career is shipped to a kid in China or a robot is built to replace you, it’s great to have a skilled job to fall back on. Everyone should start becoming more accepting of the parent working two jobs, particularly the parents who are at risk of being replaced. If we can’t set a cultural standard, people aren’t going to have as much initiative.
The new jobs you pick up don’t even need to require high levels of skill. As long as you can pick up a few extra hours at Hy-Vee, Starbucks, or even a homemade business, you might be set. Human labor is valuable. A unique personality, interesting creation, or kitsch idea will be helpful for the world. So, if you can find some way to implement them either through an online shop or making that pretty design on a mocha, it will benefit you.
Beyond this, we can increase vouchers for charter schools and loans for college students in order to build up skills earlier in life. We know this is coming. Let’s make sure it doesn’t hurt us. We were able to protect ourselves by increasing education when manufacturing was going to kill off agriculture. We see as well through the study by the Brookings Institute that education and extra skills are going to help people maintain, or even gain wages through automation. The policy we can use to reduce the harms of automation are incredibly pragmatic and within reach. Banning automation is not only outside of our pragmatic reach, but will rid us of a lot of great things.
Automation can provide a number of benefits, increase wages, and lead us to technological development that would save the economy. The real problem is making the change adaptable. We’re not talking about hyper-educating people or giving out free college. Not only is the latter option ineffective, but it goes too far. Picking up new skills and gigs can be done without a new college degree. Once we can make these very easy changes, we can largely protect ourselves from automation.
The psychological toll that we might face if we don’t address automation correctly can be a big issue for our stability. America already faces high suicide rates and drug overdoses out the window. Rather than “letting the economy figure itself out” or “leaving people to their individual desires” we have the ability through easy, pragmatic and light policy to make lives better for everyone. If Milton Friedman, one of the most well-recognized and influential economists of all time and hardcore libertarians, can support a universal basic income I think we can at least agree on some basic regulations.